365DaysOfData
A year long journey with ai from data, exploring adjacent techs
Install / Use
/learn @paudelsamir/365DaysOfDataREADME
[!note]
I'll share progress and demos on linkedin and twitter.
I won’t post daily or raw learns, updates will be for specific topics, concise, and focused on what i actually built or explored. plan is 4–5 posts per week.
This journey is about AI from scratch with data, not my entire learning history. I’ll keep building in public while also learning other adjacent techs beyond AI.
Projects Completed
| Projects | Description | Deployment | |----------|-------------|-----------| |Football Players Market Value Prediction| A 10-day end-to-end machine learning capstone project involving data scraping, cleaning, feature engineering, model training, and deployment. Achieved 94% accuracy using gradient boosting algorithms. |Live Demo 👆🏽| |Movie Recommender System|An end-to-end content-based movie recommender system leveraging a dataset of 5000 movies from Kaggle. Built with cosine similarity and TF-IDF vectorization.| Live Demo 👆🏽| |Cat vs Dog Classifier| A deep learning model leveraging VGG16 architecture, trained on an RTX 3050 Ti for 30 epochs, achieving 95% accuracy using the Kaggle Dogs vs Cats dataset. | Live Demo 👆🏽 | |Guess The Footballer By Eyes| An interactive game where users compete against AI to recognize 25 famous footballers by their eyes alone. Built with ResNet18 achieving ~70% accuracy. Features scoring system and streak tracking.| Demo 👆🏽 | |Seq2Seq Chatbot| A sequence-to-sequence chatbot trained on Cornell Movie-Dialogs Corpus using encoder-decoder architecture with Luong attention mechanism. Built from scratch in PyTorch.| Live Demo 👆🏽 | |GPT from Scratch| Complete implementation of GPT transformer architecture from scratch following Karpathy's tutorial. Includes bigram model, self-attention, multi-head attention, and complete transformer blocks.| Notebook 📓 | |Image Captioning| An end-to-end image captioning project using the Flickr8k dataset. Explored the "Show, Attend & Tell" paper, built vocabulary, extracted features with ResNet-18, and trained a transformer decoder. Achieved a BLEU-4 score of 0.18 and deployed a Streamlit demo app. | Live Demo 👆🏽 | |cineRank - A movie ranker app| Community-driven movie leaderboard app with trending picks, sentiment reviews, and personal watchlists. Built using IMDb reviews, advanced text cleaning, EDA, vectorization (BoW, TF-IDF, GloVe, BERT), and BERT fine-tuning for sentiment classification. Features leaderboard, watchlists, and real-time updates. | Live Demo 👆🏽 | |Choose Your Own Adventure| Inspired by interactive fiction like AI Dungeon, this app lets you become the protagonist in a personalized adventure story. Enter any theme—haunted mansions, space exploration, and more—and AI generates a unique branching narrative with multiple paths and endings. Features include an interactive visual map, concise story nodes (~40 words), meaningful choices, and a clean black-and-white interface for all devices. Explore different decision paths and control your own dynamic storytelling experience. | Project Demo 👆🏽 | | Projects-Based-GenAI | Hands-on GenAI projects including text generation, multimodal models, and advanced LLM fine-tuning. Explore practical implementations of state-of-the-art generative AI techniques. | Project folder | | Project-Based-AgenticAI | Applied agentic AI projects focusing on autonomous agents, multi-agent systems, and real-world agentic workflows using LangGraph and LangChain. | Project folder |
Resources
| Books & Courses | Completion Status |
|--------------------|-------------------|
| Essence of Linear Algebra @3Blue1Brown | ✅ |
| Machine Learning Specialization @Coursera | ✅ |
| Machine Learning Playlist @CampusX | ✅ |
| Hands-On Machine Learning with Scikit-Learn and TensorFlow|✅ |
| Intro to Deep Learning @MIT | ✅ |
| Deep Learning Playlist @CampusX | ✅ |
| Neural Networks @3Blue1Brown | ✅ |
| Deep Learning for Coders with fastai & PyTorch @Oreilly | ✅ |
| Neural Networks: Zero to Hero @Karpathy | ✅ |
| LangChain Intro @Pinecone | ✅ |
| 🌐 ML Resources | ✅ |
| 🌐 GenAI Handbook | ✅ |
| Deep Reinforcement Learning — Hugging Face Course | ⏳ |
| Reinforcement Learning from Human Feedback (RLHF) — Book | ⏳ |
Progress
| Days | Date | Topics | Resources | |------|--------------------|-----------------------------|--------------| | Day1 |2024‑12‑14 |Basics of Linear Algebra | 3blue1brown | | Day2 | 2024-12-15 | Decomposition, Derivation, Integration, and Gradient Descent | 3blue1brown | | Day3 | 2024-12-16 |Supervised Learning, Regression and classification|Machine Learning Specialization | | Day4 | 2024-12-17 |Unsupervised Learning: Clustering and dimensionality reduction|Machine Learning Specialization | | Day5 | 2024-12-18|Univariate linear Regression|Machine Learning Specialization | | Day6 | 2024-12-19 |Cost Functions|Machine Learning Specialization | | Day7 | 2024-12-20 |Gradient Descent|CampusX, Machine Learning Specialization | | Day8 | 2024-12-21 |Effect of learning Rate, Cost function and Data on GD|CampusX, Machine Learning Specialization | | Day9 | 2024-12-22 |Linear Regression with multiple features, Vectorization| Machine Learning Specialization | | Day10 | 2024-12-23 |Feature
